A2.2.5 Virtual Prototyping

Did you know?

What is Virtual Prototyping?

Virtual prototyping is the use of computer-based simulations and digital models to evaluate, test and communicate a design before committing to physical manufacture.

The core idea: find and fix design problems when they cost nothing to correct — on screen — rather than after expensive tooling or manufacture.

Virtual prototyping enables designers to:

• Test structural performance without breaking a physical part

• Evaluate ergonomics with a digital human before recruiting test subjects

• Present designs to clients in immersive 3D before manufacture

• Run hundreds of optimisation iterations overnight, automatically

• Reduce the total number of physical prototypes required

The Two Fundamental Model Types

Before any simulation can run, a digital model must exist. The two model types used in virtual prototyping are fundamentally different in what they describe.

Surface Models

A surface model defines the exterior geometry of an object as a shell of mathematical surfaces — like the skin of the product — with no interior, no mass, and no thickness.

How they are built

1. Designer creates curves in 3D space defining the edges of each surface patch

2. Surfaces are generated between the curves (e.g. NURBS — Non-Uniform Rational B-Splines)

3. Surfaces are trimmed and joined at their edges to form the visible exterior form

4. The result is a mathematically described shell, not a solid volume

When surface models are used

Aesthetic evaluation — surface models render and visualise with perfect fidelity, enabling designers to evaluate form, proportion, and light reflection before any engineering decisions are made. This is the dominant tool in automotive and consumer electronics styling.

CNC toolpath generation — surface geometry is sufficient to generate 5-axis CNC machining paths for physical appearance models, without the need for a fully engineered solid.

Early concept communication — clients and stakeholders can review form in a virtual environment before engineering analysis has been completed.

Limitations of surface models

No mass properties — a surface model has no volume, so mass, centre of gravity, and moment of inertia cannot be calculated from it directly.

Cannot be analysed by FEA — FEA requires a closed solid model; a surface shell must first be converted to a solid before stress analysis is possible.

Requires skill — high-quality Class A surface modelling (used in automotive) is one of the most technically demanding skills in industrial design, taking years to master.

Solid Models

A solid model defines a mathematically closed volume — a complete description of the interior and exterior of an object, including its mass, centre of gravity, and material properties.

How they are built

Most solid models are built parametrically — each feature is defined by dimensions and relationships that can be updated. Change the diameter of a boss, and all features that depend on it update automatically.

1. A 2D sketch is drawn and constrained with dimensions and geometric relationships

2. The sketch is extruded, revolved, swept or lofted to create a 3D solid feature

3. Further features (holes, fillets, chamfers, shells, ribs) are added in a logical sequence

4. The entire history of operations is recorded in a feature tree — any step can be edited

5. Assemblies combine multiple solid parts, with defined mates (constraints) between them

Why solid models are the foundation of virtual prototyping

Mass properties — the software instantly calculates mass, centre of gravity, volume, and moment of inertia for any given material density assignment.

Assembly interference checking — in a parametric assembly, the software detects where parts physically overlap (clash), identifying assembly problems before manufacture.

Drawing generation — 2D engineering drawings with dimensions and tolerances are generated directly from the solid model — no redrawing required.

FEA input — solid models provide the geometry required for finite element analysis; the solid's volume is meshed into elements for stress, thermal, and vibration analysis.

The foundation for all downstream virtual prototyping — generative design, FEA, digital human simulation, and VR visualisation all begin with a solid model.

Finite Element Analysis (FEA)

FEA is a computational simulation method that predicts how a component will respond to structural loads, heat, vibration, or fluid pressure by dividing its geometry into thousands of small elements and solving the relevant physics equations at every point.

How it works

1. A solid CAD model is imported into FEA software

2. A material is assigned (e.g. aluminium alloy 6061, with defined Young's modulus, yield strength, Poisson's ratio)

3. Boundary conditions are applied — where is the part fixed? Where are loads applied? What are the load magnitudes and directions?

4. The model is meshed — divided into a network of small elements (tetrahedra or hexahedra); finer mesh = greater accuracy but longer solve time

5. The solver calculates stress, strain, and displacement at every node in the mesh

6. Results are displayed as a colour map (stress contour plot) — red = highest stress, blue = lowest

7. The designer identifies where stress exceeds the material's yield strength, and redesigns those regions

Advantages of FEA

Finds failure points before manufacture — stress concentrations in a bracket or housing are identified digitally; the designer adds a fillet or increases wall thickness at that point before any physical part is cut.

Dramatically reduces physical testing — a design can be virtually tested against dozens of load cases overnight; only the final validated design requires physical destructive testing.

Supports lightweighting — FEA shows where material is under-stressed, revealing where designers can safely remove it — reducing product mass without compromising strength.

Accessible to small studios — SolidWorks Simulation and Fusion 360 Simulation bring FEA to studios that cannot afford industrial testing rigs.

Disadvantages of FEA

Accuracy depends on inputs — incorrect material data, poorly defined boundary conditions, or too-coarse a mesh all produce misleading results; FEA can give a false sense of certainty if inputs are not carefully validated.

Requires specialist knowledge — interpreting results, understanding mesh sensitivity, and correctly defining boundary conditions requires significant training; an untrained user can produce dangerous conclusions.

Cannot model all failure modes — standard FEA does not capture complex phenomena such as crack propagation, composite delamination, or highly non-linear plastic flow without specialist extensions.

Generative Design

Generative design is an AI-driven optimisation process where the designer defines the design space, constraints, loads, and objectives — and the software automatically generates and evaluates thousands of geometric solutions, selecting those that best meet the brief.

How it works

1. Designer defines the preserve geometry — regions that must remain (mounting holes, interfaces with other parts)

2. Designer defines the obstacle geometry — regions the material must avoid (clearance zones for adjacent parts)

3. Loads and constraints are applied (same as FEA boundary conditions)

4. Objectives are set: minimise mass, maximise stiffness, or both

5. Manufacturing method is specified — additive manufacturing, machining, casting — the algorithm only generates geometry that can be made by the chosen process

6. The software runs hundreds to thousands of FEA simulations, evolving the geometry towards the optimum

7. The designer reviews and selects from the generated alternatives, applying design judgement to the algorithmic output

Advantages of generative design

Produces solutions humans would not conceive — organic, branching structures optimised for load paths do not resemble human intuition about structural forms, yet are lighter and stronger.

Directly integrated with manufacturing constraints — specifying the manufacturing method prevents the software from generating geometry that cannot be made, saving time in design review.

Dramatically accelerates lightweighting — significant mass reductions are achieved quickly, with structural integrity validated simultaneously.

Disadvantages of generative design

Requires accurate load definition — if the loads and constraints are incorrectly defined, the generated geometry is optimised for the wrong problem and may fail in service.

Output still requires designer judgement — generative design produces geometry, not finished designs; the designer must evaluate, select, and refine outputs for aesthetics, assembly, and broader system requirements.

Computationally intensive — complex multi-load generative design runs require cloud computing and can take significant time; not all studios have access to the required software licences (Fusion 360, CATIA).

Digital Humans

Digital humans are parametric virtual human body models that simulate the dimensions, posture, movement range, vision, and reach capability of people from specified population groups, used to evaluate ergonomic design decisions.

How they work

1. The designer specifies the target user population (e.g. 5th percentile female to 95th percentile male for a European population)

2. The software generates scaled body models at multiple percentile points

3. Digital humans are placed in the product environment — seated in a cab, operating a control panel, reaching for an overhead handle

4. The software calculates joint angles and compares them to ergonomic standards (e.g. ISO 11228, RULA scores)

5. Vision analysis evaluates sightlines and blind spots from the user's eye position

6. Reach envelopes show what controls can and cannot be reached by users at each body size

7. Problem postures are identified; the designer adjusts control positions, seat heights, or grip dimensions

Advantages of digital humans

Tests the entire target population — a physical prototype can be evaluated with only a limited number of test subjects; a digital human simulation tests every percentile point in the population simultaneously.

Identifies exclusion early — if the 5th percentile female cannot reach a critical control in a digital simulation, the designer corrects it before a physical prototype is built.

Integrated with the CAD assembly — the digital human occupies the same CAD environment as the product; changes to the product geometry are immediately reflected in the ergonomic analysis.

Disadvantages of digital humans

Does not capture real human variation — a digital human is an average statistical model; it cannot represent individual variation in flexibility, strength, or movement strategies.

User experience is not simulated — comfort, perceived effort, and preference require real users; a digital human confirms dimensional adequacy, not subjective satisfaction.

Motion Capture

Motion capture (mocap) records the precise 3D movement of a human body performing a task, generating digital motion data that can be applied to digital human models, animated characters, or used directly in ergonomic analysis.

How it works

Optical systems:

1. Retroreflective markers are attached to anatomical landmarks on the subject

2. Multiple infrared cameras track marker positions simultaneously at up to 2,000 frames per second

3. Software triangulates 3D marker positions from the camera array

4. A skeletal model is fitted to the marker data

5. The resulting motion data is applied to digital human models or characters

Inertial systems:

1. Small IMU (inertial measurement unit) sensors are attached to body segments

2. Sensors measure acceleration, angular velocity, and magnetic field

3. Sensor fusion algorithms compute joint angles and segment orientations

4. No camera infrastructure required — can be used in any environment

How designers use motion capture

Ergonomic analysis — assembly workers' postures during manufacturing tasks are captured and analysed for injury risk, informing workstation redesign.

Product interaction studies — how users actually grip, operate, and move a product is captured and compared to intended design assumptions.

Animation and visualisation — natural human movement is applied to digital characters in product animations, marketing films, and VR environments.

Sports product development — athletes' movement patterns during sport-specific actions are captured to inform the biomechanical requirements of footwear, protective equipment, and prosthetics.

Limitations of motion capture

Controlled environment required — optical systems cannot be used outdoors in sunlight; inertial systems suffer from magnetic field interference in industrial environments.

Reflects current behaviour, not optimised behaviour — capturing how someone currently performs a task identifies problems but does not automatically identify the best solution.

Haptic Technology

Haptic technology uses force feedback devices to allow designers to physically feel virtual surfaces, materials, and resistances as they interact with a digital model — adding the sense of touch to virtual prototyping.

How it works

1. A designer holds or wears a haptic device (e.g. a stylus, glove, or exoskeleton)

2. As the designer moves the device in physical space, sensors track its position in 3D

3. The software checks whether the device has penetrated a virtual surface

4. If contact is detected, motors in the device apply a counterforce to the designer's hand, simulating the feel of the virtual surface

5. The stiffness, texture, and resistance can be varied digitally to simulate different materials

How designers use haptics

Virtual clay modelling — designers sculpt organic product forms (handles, consumer goods casings, automotive interiors) by hand, feeling resistance as they push and pull virtual material — combining the intuitive quality of physical clay modelling with the editability of digital files.

Medical device and surgical tool design — surgeons and designers jointly evaluate grip, balance, and tool feel in a virtual environment before physical prototypes are manufactured.

Assembly feasibility — operators feel the resistance of virtual snap fits, torque requirements, and insertion forces in a simulated assembly, flagging assembly quality problems early.

Limitations of haptics

Force fidelity is limited — current haptic devices can simulate surface contact and gross stiffness, but cannot yet accurately reproduce fine surface texture, temperature, or the full force range of real material interactions.

High cost of high-fidelity systems — industrial-grade haptic arms cost £10,000–30,000; consumer devices have very limited force output.

Virtual Reality (VR) and Augmented Reality (AR)

How VR is used in virtual prototyping

Full-scale design review — a digital model is experienced at 1:1 scale in VR; designers and engineers walk around the virtual product, open virtual doors, and sit in virtual seating positions — identifying scale and proportion issues that are invisible on a 2D screen.

Client presentation — stakeholders who cannot read engineering drawings experience the product in an immersive, intuitive environment before manufacture — reducing the risk of late-stage change requests.

Interior environment evaluation — automotive, aircraft, and architectural interiors are evaluated for spatial quality, lighting, material appearance, and outward visibility before any physical interior is built.

Multi-site collaborative review — distributed teams in different countries occupy the same virtual design space simultaneously, reviewing and marking up the model in real time.

How AR is used in virtual prototyping

In-context product placement — a digital product model is overlaid onto the real physical environment where it will be used; the designer evaluates whether the product fits visually and spatially into its real context.

Assembly guidance — digital assembly instructions are overlaid directly onto a physical prototype or sub-assembly, guiding technicians through complex build sequences.

Size and proportion validation — a new product is visualised at 1:1 scale in the real space (kitchen, bathroom, vehicle interior) before physical production, allowing fast client approval.

Advantages of VR/AR

Communicates scale that 2D screens cannot — the single biggest advantage; human perception of scale, proportion, and spatial quality requires physical experience that a monitor cannot provide.

Dramatically reduces costly late-stage changes — design errors in spatial quality, ergonomics, and proportion discovered in VR cost nothing to fix; the same errors discovered after physical manufacture are expensive.

Democratises design review — stakeholders who are not trained in reading CAD models or engineering drawings can meaningfully participate in design decisions through VR/AR.

Disadvantages of VR/AR

Hardware cost and setup — high-fidelity VR systems with tracking require significant hardware investment and dedicated physical space; headset-induced discomfort (motion sickness, headset weight) limits session duration.

Visual fidelity gap — real-time rendering in VR does not yet match the photorealism of offline rendering; subtle surface quality, material finish, and lighting nuances may be misread.

CAD pipeline complexity — converting a detailed engineering CAD assembly into a real-time VR-compatible format requires polygon reduction, material reprocessing, and specialist software (e.g. PTC Creo Illustrate, Autodesk VRED, Unity, Unreal Engine).

Why Designers Use Virtual Prototypes — Summary

The Relationship Between Virtual and Physical Prototyping

Virtual and physical prototyping are complementary, not competing processes:

• Virtual prototyping removes the cost and time of physical iteration for structural, ergonomic, and scale evaluation

• Physical prototyping remains essential for evaluating real material behaviour, surface finish quality, user feel, and manufacturing feasibility

• The best design development process uses virtual prototyping early and often to reduce the number of physical prototypes required, then uses physical prototypes to validate what the virtual model cannot simulate

Key Vocabulary

Practice Questions

Question 1

State the difference between a surface model and a solid model. [2]

Question 2 

Describe how Finite Element Analysis (FEA) is used in virtual prototyping. [4]

Question 3  A medical device company is developing a new surgical hand tool. Describe how they could use three different virtual prototyping technologies during development, justifying each choice in the context of a surgical hand tool. [6]

Sources

• International Baccalaureate Organization. (2023). Design Technology Guide: First Assessment 2027. IBO, Geneva. — The primary syllabus document defining A2.2.5 content requirements and assessment objectives.

• Hallgrimsson, B. (2012). Prototyping and Modelmaking for Product Design. Laurence King Publishing. — Covers virtual and physical prototyping workflows in an industrial design context; directly relevant to surface modelling and prototype decision-making.

• Ulrich, K. & Eppinger, S. (2015). Product Design and Development (6th ed.). McGraw-Hill. — Standard engineering design textbook covering FEA, CAD modelling, and prototype strategy in the product development process.

• Lemons, G., Carberry, A., Swan, C., Jarvin, L. & Rogers, C. (2010). The Benefits of Model Building in Teaching Engineering Design. Design Studies, 31(3), 288–309. —

• Moaveni, S. (2011). Finite Element Analysis: Theory and Application with ANSYS (4th ed.). Pearson. — Definitive reference for FEA theory, meshing, boundary conditions, and results interpretation.

• Autodesk. (2024). Fusion 360 Simulation Documentation. Autodesk Knowledge Network. Retrieved from: knowledge.autodesk.com — Software documentation for FEA workflows accessible to students and small studios.

• Dassault Systèmes. (2024). SolidWorks Simulation User Guide. Dassault Systèmes. — Documents the FEA workflow used in the most widely taught CAD/FEA platform in design education.

• Autodesk Research. (2016). The Airbus Bionic Partition Project. Autodesk. — Industry case study documenting the 45% mass reduction achieved using generative design on the A320 cabin partition; directly cited in this document.

• Du Bois, P., Zouein, P. & Issa, R. (2021). Generative Design: Upstream Integration into the Design Process. Journal of Computational Design and Engineering, 8(4), 1103–1117. — Academic source covering the workflow and limitations of generative design in professional practice.

• Autodesk. (2024). Generative Design in Fusion 360: Learning Resources. Autodesk University. Retrieved from: autodesk.com/autodesk-university — Accessible tutorials and case studies suitable for student independent study.

• Chaffin, D.B. (2007). Human Motion Simulation for Vehicle and Workplace Design. Human Factors and Ergonomics in Manufacturing, 17(5), 475–484. — Foundational academic paper on digital human modelling methodology and its application to product and workspace design.

• Siemens. (2024). Tecnomatix Jack Human Simulation. Siemens Digital Industries Software. Retrieved from: plm.automation.siemens.com — Software documentation for the JACK digital human tool referenced in the technology section.

• McAtamney, L. & Corlett, E.N. (1993). RULA: A Survey Method for the Investigation of Work-Related Upper Limb Disorders. Applied Ergonomics, 24(2), 91–99. — Original publication defining RULA, the ergonomic scoring system referenced throughout the digital humans section.

• Vicon. (2024). Motion Capture Technology: Applications in Product Development. Vicon Motion Systems. Retrieved from: vicon.com — Industry documentation for optical motion capture systems referenced in this document.

• Roetenberg, D., Luinge, H. & Slycke, P. (2009). Xsens MVN: Full 6DOF Human Motion Tracking Using Miniature Inertial Sensors. Xsens Technologies. — Technical paper documenting the inertial motion capture methodology underlying the Xsens system referenced in the technology section.

• Salisbury, K., Conti, F. & Barbagli, F. (2004). Haptic Rendering: Introductory Concepts. IEEE Computer Graphics and Applications, 24(2), 24–32. — Foundational review of haptic rendering principles, force feedback, and fidelity limitations directly relevant to this section.

• 3D Systems. (2024). Geomagic Touch Haptic Device Documentation. 3D Systems. Retrieved from: 3dsystems.com — Product documentation for the haptic device referenced in the technology section.

• Bryson, S. (1996). Virtual Reality in Scientific Visualisation. Communications of the ACM, 39(5), 62–71. — Early foundational paper establishing VR as a tool for design evaluation and spatial comprehension.

• Microsoft. (2024). HoloLens 2 for Industry: Design Review Applications. Microsoft Mixed Reality. Retrieved from: microsoft.com/en-us/hololens — Industry documentation for the AR platform referenced in this document, with specific design review case studies.

• Autodesk. (2024). VRED Professional: VR Design Review Workflows. Autodesk. Retrieved from: autodesk.com/products/vred — Software documentation for the automotive VR design review platform referenced in the CAD pipeline discussion.

• Milgram, P. & Kishino, F. (1994). A Taxonomy of Mixed Reality Visual Displays. IEICE Transactions on Information and Systems, 77(12), 1321–1329. — The original academic paper defining the Reality–Virtuality Continuum underpinning the VR/AR distinction used in this document.

• ISO 11228-1:2021. Ergonomics — Manual Handling — Lifting, Lowering and Carrying. International Organization for Standardization. — The ergonomic standard referenced in the digital humans section for joint angle and posture evaluation.

• ISO 9241-210:2019. Ergonomics of Human-System Interaction — Human-Centred Design for Interactive Systems. International Organization for Standardization. — The human-centred design process standard underpinning the use of digital humans and motion capture in design evaluation.

• Autodesk Design Academy. (2024). Free CAD and Simulation Learning Resources for Students. Retrieved from: autodesk.com/education — Free access to Fusion 360 with FEA and generative design for students; directly supports practical application of A2.2.5 content.

• GrabCAD Community. (2024). CAD Model Library and Engineering Tutorials. Retrieved from: grabcad.com — Free parametric CAD models and tutorials supporting independent study of solid modelling workflows.

• Simscale. (2024). Free Cloud FEA and CFD for Students. Retrieved from: simscale.com — Browser-based FEA and CFD platform; no software installation required; free student accounts available.